A quality inspection method and system based on 3D printing technology

Abstract

The present invention discloses a quality detection method based on 3D printing technology. At a plurality of preset positions in the product generation area of a 3D printing device, a monocular camera device is set. Two-dimensional images of the product at the preset positions are collected through the monocular camera device. The moment of collecting the two-dimensional images of the 3D product is a preset moment during the three-dimensional printing process. 3D printing model data is obtained, and two-dimensional model data corresponding to the preset positions is obtained according to the preset parameters of the preset plurality of positions. The preset parameters include angles and distances. Different quality weight distributions are performed on the collected images at the preset plurality of positions, and finally, the two-dimensional images corresponding to the same positions of the collected two-dimensional images at all the preset positions and the 3D model data are compared. Weighted processing is performed on all positions according to the comparison results to obtain the quality score of the final product. When the quality score is less than a preset value, it is determined that the 3D printed product is unqualified.

Description

Technical Field

[0001] The present invention relates to the technical field of 3D printing, and in particular to a quality detection method and system based on 3D printing technology. Background Art

[0002] A 3D printer, also known as a three-dimensional printer, is called additive manufacturing technology. It is a machine that uses rapid prototyping technology. Based on a digital model file, it uses a forming material and constructs a three-dimensional entity by printing layer by layer. Before printing, it is necessary to use computer modeling software to model to form a 3D model to be printed, and then "partition" the built 3D model into cross-sections layer by layer, that is, slicing, so as to guide the 3D printer to print layer by layer. 3D printers have been widely used in product manufacturing. The working principle of 3D printers is basically the same as that of traditional printers, consisting of a control component, a mechanical component, a print head, consumables (i.e., forming materials) and a medium, etc., and the printing principle is also basically similar.

[0003] The existing product quality analysis and control system of 3D printers cannot intuitively show the production quality situation to the staff, and the quality supervision method is more through the supervision of the staff or by using a complex three-dimensional image coordinate comparison method for similarity comparison. For example, the Chinese patent CN201710886339.X in the prior art discloses a 3D printing product quality detection and repair method combining a three-dimensional model and machine vision. Among them, the quality detection method of the printed product is detailed as "100. Evaluate the quality of the 3D printed product. First, scan the 3D printed product to obtain the corresponding point cloud data of the product, and align the STL model corresponding to the printed product with the scanned point cloud to obtain the quality evaluation and the error between the model and the point cloud data; 200. Initialize the product position, that is, determine the world coordinates corresponding to the mechanical system where the printed product is placed; 300. Align the world coordinates with the model coordinates, and map the error between the model and the point cloud data to the world coordinates; 400. According to the obtained world coordinates and the corresponding error, visually display the error distribution between the product model and the point cloud data, and select corresponding path planning for different 3D printed products to achieve repair", that is, the quality detection of existing 3D printed products uses the evaluation and comparison of point cloud data, and uses relatively complex data conversion and processing, which is not conducive to quality supervision during mass printing. Therefore, the present invention aims to provide a rapid 3D product evaluation and quality detection method based on machine vision. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art. For this reason, the present invention discloses a quality detection method based on 3D printing technology, and the quality detection method includes the following steps:

[0005] Step 1: At a plurality of preset positions in the product generation area of the 3D printing device, set up monocular camera devices, and collect two-dimensional images of the product at the preset positions through the monocular camera devices. Among them, the moment of collecting the two-dimensional images of the 3D product is a preset moment during the three-dimensional printing process;

[0006] Step 2: Obtain 3D printing model data, and obtain two-dimensional model data corresponding to the preset positions according to the preset parameters of the plurality of preset positions. Among them, the preset parameters include angle and distance;

[0007] Step 3: Perform different quality weight assignments on the collected images at the preset plurality of positions, and finally evaluate the comparison between the two-dimensional images collected at all preset positions and the two-dimensional images corresponding to the same positions of the 3D model data;

[0008] Step 4: Perform weighted processing on all positions according to the comparison results to obtain the quality score of the final product. When the quality score is less than the preset value, it is determined that the 3D printed product is unqualified.

[0009] Furthermore, the preset plurality of positions are three positions, that is, the three positions corresponding to the perspective directions of the three views of the 3D printed product, and the distance from the monocular camera device to the printed product is the pre-installed distance.

[0010] Furthermore, when the number of the preset plurality of positions is 3, the calculation expression of the final quality score Q is as follows:

[0011] Q = q1×k1 + q2×k2 + q3×k3

[0012] Among them, q1 is the quality score of the first preset position, and k1 is the weight corresponding to the first preset position; q2 is the quality score of the second preset position, and k2 is the weight corresponding to the second preset position; q3 is the quality score of the third preset position, and k3 is the weight corresponding to the third preset position.

[0013] Furthermore, the calculation method of the quality scores at different positions is to perform grayscale processing on the collected images, then perform same-size scaling on the collected two-dimensional images and the two-dimensional images corresponding to the same positions of the generated 3D model data, and then calculate the cosine similarity between the two. The calculated cosine similarity is the quality score of this position.

[0014] Furthermore, the weights assigned to different positions are the weight values preset by the administrator, which reflect the importance of different positions in the quality assessment process.

[0015] From a hardware perspective, the present invention also discloses a quality detection system based on 3D printing technology. The quality detection system includes the following functional modules:

[0016] An image acquisition module sets a monocular camera device at a plurality of preset positions in the product generation area of a 3D printing device, and acquires two-dimensional images of the product at the preset positions through the monocular camera device. Among them, the moment of acquiring the two-dimensional images of the 3D product is a preset moment during the three-dimensional printing process;

[0017] A model data processing module obtains 3D printing model data, and obtains two-dimensional model data corresponding to the preset positions according to the preset parameters of the preset plurality of positions, where the preset parameters include angles and distances;

[0018] A quality calculation module performs different quality weight assignments on the acquired images at a plurality of preset positions, and finally compares the two-dimensional images acquired at all preset positions with the two-dimensional images corresponding to the same positions of the 3D model data;

[0019] A quality evaluation module performs weighted processing on all positions according to the comparison results to obtain the quality score of the final product. When the quality score is less than a preset value, it is determined that the 3D printed product is unqualified.

[0020] Furthermore, the image acquisition module further includes: the preset plurality of positions are three positions, namely the three positions corresponding to the three-view perspective directions of the 3D printed product, and the distance from the monocular camera device to the printed product is a pre-installed distance.

[0021] Furthermore, the quality evaluation module further includes that when the number of preset positions is 3, the calculation expression of the final quality score Q is as follows:

[0022] Q = q1×k1 + q2×k2 + q3×k3

[0023] Wherein, q1 is the quality score of the first preset position, and k1 is the weight corresponding to the first preset position; q2 is the quality score of the second preset position, and k2 is the weight corresponding to the second preset position; q3 is the quality score of the third preset position, and k3 is the weight corresponding to the third preset position.

[0024] Furthermore, the calculation method of the quality scores at different positions is to perform grayscale processing on the acquired images, then perform same-size scaling on the two-dimensional images acquired and the two-dimensional images corresponding to the same positions of the generated 3D model data, and then calculate the cosine similarity between the two. The calculated cosine similarity is the quality score of this position.

[0025] Furthermore, the weights assigned to different positions are weight values preset by the administrator, which reflect the importance of different positions in the quality evaluation process.

[0026] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention realizes the effect detection of 3D printed products by using the method of two-dimensional image comparison from multiple angles. In the present invention, products at different printing stages can be photographed, different quality weight distributions are performed on the collected images from different angles, and finally, a comprehensive evaluation is carried out on the comparison between the two-dimensional images collected at all preset angles and the two-dimensional images corresponding to the same angles of the 3D model data. In this way, the duration of data conversion is greatly reduced, and the comparison efficiency is improved, which is suitable for high-speed or large-scale three-dimensional printing work operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The present invention can be further understood from the following description in conjunction with the drawings. The components in the drawings are not necessarily drawn to scale, but the emphasis is placed on showing the principles of the embodiments. In the drawings, the same reference numerals designate corresponding parts in different views.

[0028] Figure 1 It is a flowchart of a quality detection method for a stereolithography 3D printer of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0029] As Figure 1 shown, a quality detection method based on 3D printing technology, the quality detection method includes the following steps:

[0030] Step 1, at a plurality of preset positions in the product generation area of the 3D printing device, set a monocular camera device, and collect two-dimensional images of the product at the preset positions through the monocular camera device. Among them, the moment of collecting the two-dimensional images of the 3D product is a preset moment during the three-dimensional printing process;

[0031] Step 2, obtain 3D printing model data, and obtain two-dimensional model data corresponding to the preset positions according to the preset parameters of the plurality of preset positions, where the preset parameters include angles and distances;

[0032] Step 3, perform different quality weight distributions on the collected images at the plurality of preset positions, and finally evaluate the comparison between the two-dimensional images collected at all preset positions and the two-dimensional images corresponding to the same positions of the 3D model data;

[0033] Step 4, perform weighted processing on all positions according to the comparison results to obtain the quality score of the final product. When the quality score is less than the preset value, it is determined that the 3D printed product is unqualified.

[0034] Furthermore, the plurality of preset positions are three positions, that is, the three positions corresponding to the three-view perspective directions of the 3D printed product, and the distance from the monocular camera device to the printed product is a pre-installed distance.

[0035] Further, when the number of preset positions is 3, the calculation expression of the final quality score Q is as follows:

[0036] Q = q1 × k1 + q2 × k2 + q3 × k3

[0037] Among them, q1 is the quality score of the first preset position, k1 is the weight corresponding to the first preset position; q2 is the quality score of the second preset position, k2 is the weight corresponding to the second preset position; q3 is the quality score of the third preset position, k3 is the weight corresponding to the third preset position.

[0038] Further, the calculation method of the quality score at different positions is to grayscale the collected image, then scale the same-position corresponding two-dimensional images of the collected two-dimensional image and the generated 3D model data to the same size, and then calculate the cosine similarity between the two. The calculated cosine similarity is the quality score of this position.

[0039] Further, the weights assigned to different positions are the weight values preset by the manager, which reflect the importance of different positions in the quality assessment process.

[0040] From a hardware perspective, the present invention also discloses a quality detection system based on 3D printing technology. The quality detection system includes the following functional modules:

[0041] An image acquisition module, which sets a monocular camera device at a preset plurality of positions in the product generation area of the 3D printing device, and acquires the two-dimensional image of the product at the preset positions through the monocular camera device. Among them, the moment of acquiring the two-dimensional image of the 3D product is a preset moment during the three-dimensional printing process;

[0042] A model data processing module, which acquires 3D printing model data, and acquires two-dimensional model data corresponding to the preset positions according to the preset parameters of the preset plurality of positions. Among them, the preset parameters include angle and distance;

[0043] A quality calculation module, which performs different quality weight assignments on the acquired images at the preset plurality of positions, and finally compares the two-dimensional images acquired at all the preset positions with the two-dimensional images corresponding to the same positions of the 3D model data;

[0044] A quality evaluation module, which performs weighted processing on all positions according to the comparison results to obtain the quality score of the final product. When the quality score is less than the preset value, it is determined that the 3D printed product is unqualified.

[0045] Further, the image acquisition module further includes: the preset multiple positions are three positions, namely the three positions corresponding to the perspective directions of the three views of the 3D printed product, and the distance from the monocular camera device to the printed product is the pre-installed distance.

[0046] Further, the quality assessment module further includes that when the number of preset multiple positions is 3, the calculation expression of the final quality score Q is as follows:

[0047] Q = q1×k1 + q2×k2 + q3×k3

[0048] Wherein, q1 is the quality score of the first preset position, k1 is the weight corresponding to the first preset position; q2 is the quality score of the second preset position, k2 is the weight corresponding to the second preset position; q3 is the quality score of the third preset position, k3 is the weight corresponding to the third preset position.

[0049] Further, the method for calculating the quality scores at different positions is to perform grayscale processing on the acquired images, then perform same-size scaling on the same positions of the acquired two-dimensional images and the two-dimensional images corresponding to the generated 3D model data, and then calculate the cosine similarity between the two. The calculated cosine similarity is the quality score of that position.

[0050] Further, the weights assigned to different positions are the weight values preset by the administrator, which reflect the importance of different positions in the quality assessment process.

[0051] In this embodiment, the preferred method for calculating the quality scores of the images of the 3D printed product acquired at different positions is:

[0052]

[0053]

[0054] It should also be noted that the terms "include", "comprise" or any other variation thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, commodity or device including the said element.

[0055] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0056] Although the present invention has been described above with reference to various embodiments, it should be understood that many changes and modifications can be made without departing from the scope of the present invention. Therefore, it is intended that the above detailed description be regarded as illustrative rather than restrictive, and it should be understood that the following claims (including all equivalents) are intended to define the spirit and scope of the present invention. These embodiments should be understood as only for illustrating the present invention and not for limiting the protection scope of the present invention. After reading the content described in the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.

Claims

1. A quality inspection method based on 3D printing technology, characterized in that, The quality inspection method includes the following steps: Step 1: Set up monocular camera devices at a plurality of preset positions in the product generation area of the 3D printing device, and collect two-dimensional images of the product at the preset positions through the monocular camera devices. Among them, the moment of collecting the two-dimensional images of the 3D product is a preset moment during the three-dimensional printing process; the preset plurality of positions are three positions, namely the three positions corresponding to the three-view perspective directions of the 3D printed product, and the distance from the monocular camera device to the printed product is the pre-installed distance. Step 2: Obtain the 3D printing model data, and obtain the two-dimensional model data corresponding to the preset positions according to the preset parameters of the preset plurality of positions, where the preset parameters include angle and distance. Step 3: Perform different quality weight assignments on the collected images at the preset plurality of positions, and finally evaluate the comparison between the two-dimensional images collected at all preset positions and the two-dimensional images corresponding to the same positions of the 3D model data. Step 4: Perform weighted processing on all positions according to the comparison results to obtain the quality score of the final product. When the quality score is less than the preset value, it is determined that the 3D printed product is unqualified. When the number of preset positions is 3, the calculation expression of the final quality score Q is as follows: Q = q1×k1 + q2×k2 + q3×k3 Among them, q1 is the quality score of the first preset position, and k1 is the weight corresponding to the first preset position; q2 is the quality score of the second preset position, and k2 is the weight corresponding to the second preset position; q3 is the quality score of the third preset position, and k3 is the weight corresponding to the third preset position; among them, the calculation method of the quality scores at different positions is to perform grayscale processing on the collected images, then perform same-size scaling on the two-dimensional images collected and the two-dimensional images corresponding to the same positions of the generated 3D model data, and then calculate the cosine similarity between the two. The calculated cosine similarity is the quality score of this position.

2. The quality inspection method based on 3D printing technology according to claim 1, characterized in that, The weights assigned to different positions are the weight values preset by the manager, which reflect the importance of different positions in the quality assessment process.

3. A quality inspection system based on 3D printing technology, characterized in that, The quality inspection system includes the following functional modules: Image acquisition module: Set up monocular camera devices at a plurality of preset positions in the product generation area of the 3D printing device, and collect two-dimensional images of the product at the preset positions through the monocular camera devices. Among them, the moment of collecting the two-dimensional images of the 3D product is a preset moment during the three-dimensional printing process; the preset plurality of positions are three positions, namely the three positions corresponding to the three-view perspective directions of the 3D printed product, and the distance from the monocular camera device to the printed product is the pre-installed distance. Model data processing module: Obtain the 3D printing model data, and obtain the two-dimensional model data corresponding to the preset positions according to the preset parameters of the preset plurality of positions, where the preset parameters include angle and distance. Quality calculation module: Perform different quality weight assignments on the collected images at the preset plurality of positions, and finally evaluate the comparison between the two-dimensional images collected at all preset positions and the two-dimensional images corresponding to the same positions of the 3D model data. The quality assessment module performs weighted processing on all positions according to the comparison result to obtain the quality score of the final product. When the quality score is less than the preset value, the 3D printed product is judged to be unqualified. When the number of preset positions is 3, the calculation expression of the final quality score Q is as follows: Q = q1×k1 + q2×k2 + q3×k3 Where, q1 is the quality score of the first preset position, and k1 is the weight corresponding to the first preset position; q2 is the quality score of the second preset position, and k2 is the weight corresponding to the second preset position; q3 is the quality score of the third preset position, and k3 is the weight corresponding to the third preset position. The calculation method of the quality score at different positions is to perform grayscale processing on the collected image, then perform same-size scaling on the same-position corresponding two-dimensional images of the collected two-dimensional image and the generated 3D model data, and then calculate the cosine similarity between the two. The calculated cosine similarity is the quality score of this position.

4. A quality inspection system based on 3D printing technology according to claim 3, characterized in that, The weights assigned to different positions are the weight values preset by the manager, which reflect the importance of different positions in the quality assessment process.

Images